OpenAI Computer Use Agent

Integrate OpenAI CUA with Browserbase for scalable browser agents

This guide walks you through integrating OpenAI’s Computer Use Agent (CUA) with Browserbase for scalable cloud browser agents.

CUA is an AI model that can see the screen, understand context, and take actions within a browser — enabling advanced interaction with web applications. Pair CUA with Browserbase’s scalable remote browser infrastructure to run AI-powered browser agents effortlessly in the cloud.

Try out the Computer Use Agent now: cua.browserbase.com

Prerequisites

• OpenAI API key with Computer Use Agent access
• Browserbase account and API key
• Python 3.8+

Basic integration

This basic setup will get you up and running with a CUA agent using Browserbase as the underlying browser automation platform.

1. Clone the repository

   git clone https://github.com/openai/openai-cua-sample-app.git
   
2. Install the required packages

   pip install -r "requirements.txt"
   
3. Set the environment variables

   BROWSERBASE_API_KEY=YOUR_API_KEY
   OPENAI_API_KEY=YOUR_OPENAI_API_KEY
   OPENAI_ORG=YOUR_OPENAI_ORG
   
4. Run the agent

   Update the prompt in your cli to change the behavior of the agent

   python cli.py --computer browserbase --input "go to hackernews, tell me the top news"
   

Customizing the CUA agent

You can customize the CUA agent by updating the flags in the CLI:

• --input: The initial input to the agent (optional: the CLI will prompt you for input if not provided)
• --debug: Enable debug mode.
• --show: Show images (screenshots) during the execution.
• --start-url: Start the browsing session with a specific URL (only for browser environments). By default, the CLI will start the browsing session with https://bing.com.

Related resources

Browserbase documentation

Explore the full Browserbase documentation

OpenAI API documentation

Learn more about OpenAI’s APIs

Example projects on GitHub

Find the sample project for this integration